We describe a novel deep learning neural network method and its application to impute assay pIC 50 values. Unlike conventional machine learning approaches, this method is trained on sparse bioactivity data as input, typical of that found in public and commercial databases, enabling it to learn directly from correlations between activities measured in dierent assays.In two case studies on public domain data sets we show that the neural network method outperforms traditional quantitative structureactivity relationship (QSAR) models and other leading approaches. Furthermore, by focussing on only the most condent predictions the accuracy is increased to R 2 > 0.9 using our method, as compared to R 2 = 0.44 when reporting all predictions.
Contemporary
deep learning approaches still struggle to bring a
useful improvement in the field of drug discovery because of the challenges
of sparse, noisy, and heterogeneous data that are typically encountered
in this context. We use a state-of-the-art deep learning method, Alchemite,
to impute data from drug discovery projects, including multitarget
biochemical activities, phenotypic activities in cell-based assays,
and a variety of absorption, distribution, metabolism, and excretion
(ADME) endpoints. The resulting model gives excellent predictions
for activity and ADME endpoints, offering an average increase in R
2 of 0.22 versus quantitative structure–activity
relationship methods. The model accuracy is robust to combining data
across uncorrelated endpoints and projects with different chemical
spaces, enabling a single model to be trained for all compounds and
endpoints. We demonstrate improvements in accuracy on the latest chemistry
and data when updating models with new data as an ongoing medicinal
chemistry project progresses.
We propose a smooth pseudopotential for the contact interaction acting between ultracold atoms confined to two dimensions. The pseudopotential reproduces the scattering properties of the repulsive contact interaction up to 200 times more accurately than a hard disk potential, and in the attractive branch gives a 10-fold improvement in accuracy over the square well potential. Furthermore, the new potential enables diffusion Monte Carlo simulations of the ultracold gas to be run 15 times quicker than was previously possible.
Weak attractive interactions in a spin-imbalanced Fermi gas induce a multi-particle instability, binding multiple fermions together. The maximum binding energy per particle is achieved when the ratio of the number of up-and down-spin particles in the instability is equal to the ratio of the up-and down-spin densities of states in momentum at the Fermi surfaces, to utilize the variational freedom of all available momentum states. We derive this result using an analytical approach, and verify it using exact diagonalization. The multi-particle instability extends the Cooper pairing instability of balanced Fermi gases to the imbalanced case, and could form the basis of a many-body state, analogously to the construction of the Bardeen-Cooper-Schrieffer theory of superconductivity out of Cooper pairs. c † ↑(q−k) c † ↓k c ↓k c ↑(q−k ) , (1) arXiv:1712.09847v1 [cond-mat.str-el]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.